Image Codec Research Articles

NN-VVC: A Hybrid Learned-Conventional Video Codec Targeting Humans and Machines

Advancements in artificial intelligence have significantly increased the use of images and videos in machine analysis algorithms, predominantly neural networks. However, the traditional methods of compressing, storing and transmitting media have been optimized for human viewers rather than machines. Current research in coding images and videos for machine analysis has evolved in two distinct paths. The first is characterized by End-to-End (E2E) learned codes, which show promising results in image coding but have yet to match the performance of leading Conventional Video Codecs (CVC) and suffer from a lack of interoperability. The second path optimizes CVC, such as the Versatile Video Coding (VVC) standard, for machine-oriented reconstruction. Although CVC-based approaches enjoy widespread hardware and software compatibility and interoperability, they often fall short in machine task performance, especially at lower bitrates. This paper proposes a novel hybrid codec for machines named NN-VVC, which combines the advantages of an E2E-learned image codec and a CVC to achieve high performance in both image and video coding for machines. Our experiments show that the proposed system achieved up to − 43.20% and − 26.8% Bjøntegaard Delta rate reduction over VVC for image and video data, respectively, when evaluated on multiple different datasets and machine vision tasks according to the common test conditions designed by the VCM study group in MPEG standardization activities. Furthermore, to improve reconstruction quality, we introduce a human-focused branch into our codec, enhancing the visual appeal of reconstructions intended for human supervision of the machine-oriented main branch.

Design and Implementation of SoC-based Image Codec IP Cores

Design and Implementation of SoC-based Image Codec IP Cores

A Low-Bit-Rate Image Semantic Communication System Based on Semantic Graph

In the progress of research in the field of semantic communication, most efforts have been focused on optimizing the signal-to-noise ratio (SNR), while the design and optimization of the bit rate required for transmission have been relatively neglected. To address this issue, this study introduces an innovative low-bit-rate image semantic communication system model, which aims to reconstruct images through the exchange of semantic information rather than traditional symbol transmission. This model employs an image feature extraction and optimization reconstruction framework, achieving visually satisfactory and semantically consistent reconstruction performance at extremely low bit rates (below 0.03 bits per pixel (bpp)). Unlike previous methods that used pixel accuracy as the standard for distortion measurement, this research introduces multiple perceptual metrics to train and evaluate the proposed image semantic encoding model, aligning more closely with the fundamental purpose of semantic communication. Experimental results demonstrate that, compared to WebP, JPEG, and deep learning-based image codecs, our model not only provides a more visually pleasing reconstruction effect but also significantly reduces the required bit rate while maintaining semantic consistency.

Deep Network for Image Compressed Sensing Coding Using Local Structural Sampling

Existing image compressed sensing (CS) coding frameworks usually solve an inverse problem based on measurement coding and optimization-based image reconstruction, which still exist the following two challenges: (1) the widely used random sampling matrix, such as the Gaussian Random Matrix (GRM), usually leads to low measurement coding efficiency, and (2) the optimization-based reconstruction methods generally maintain a much higher computational complexity. In this article, we propose a new convolutional neural network based image CS coding framework using local structural sampling (dubbed CSCNet) that includes three functional modules: local structural sampling, measurement coding, and Laplacian pyramid reconstruction. In the proposed framework, instead of GRM, a new local structural sampling matrix is first developed, which is able to enhance the correlation between the measurements through a local perceptual sampling strategy. Besides, the designed local structural sampling matrix can be jointly optimized with the other functional modules during the training process. After sampling, the measurements with high correlations are produced, which are then coded into final bitstreams by the third-party image codec. Last, a Laplacian pyramid reconstruction network is proposed to efficiently recover the target image from the measurement domain to the image domain. Extensive experimental results demonstrate that the proposed scheme outperforms the existing state-of-the-art CS coding methods while maintaining fast computational speed.

NAWQ-SR: A Hybrid-Precision NPU Engine for Efficient On-Device Super-Resolution

In recent years, image and video delivery systems have begun integrating deep learning super-resolution (SR) approaches, leveraging their unprecedented visual enhancement capabilities while reducing reliance on networking conditions. Nevertheless, deploying these solutions on mobile devices still remains an active challenge as SR models are excessively demanding with respect to workload and memory footprint. Despite recent progress on on-device SR frameworks, existing systems either penalize visual quality, lead to excessive energy consumption or make inefficient use of the available resources. This work presents NAWQ-SR, a novel framework for the efficient on-device execution of SR models. Through a novel hybrid-precision quantization technique and a runtime neural image codec, NAWQ-SR exploits the multi-precision capabilities of modern mobile NPUs in order to minimize latency, while meeting user-specified quality constraints. Moreover, NAWQ-SR selectively adapts the arithmetic precision at run time to equip the SR DNN's layers with wider representational power, improving visual quality beyond what was previously possible on NPUs. Altogether, NAWQ-SR achieves an average speedup of 7.9×, 3× and 1.91× over the state-of-the-art on-device SR systems that use heterogeneous processors (MobiSR), CPU (SplitSR) and NPU (XLSR), respectively. Furthermore, NAWQ-SR delivers an average of 3.2× speedup and 0.39 dB higher PSNR over status-quo INT8 NPU designs, but most importantly mitigates the negative effects of quantization on visual quality, setting a new state-of-the-art in the attainable quality of NPU-based SR.

Graph Based Cross-Channel Transform for Color Image Compression

Adaptive transform coding is gaining more and more attention for better mining of image content over fixed transforms such as discrete cosine transform (DCT). As a special case, graph transform learning establishes a novel paradigm for the graph-based transforms. However, there still exists a challenge for graph transform learning-based image codecs design on natural image compression, and graph representation cannot describe regular image samples well over graph-structured data. Therefore, in this article, we propose a cross-channel graph-based transform (CCGBT) for natural color image compression. We observe that neighboring pixels having similar intensities should have similar values in the chroma channels, which means that the prominent structure of the luminance channel is related to the contours of the chrominance channels. A collaborative design of the learned graphs and their corresponding distinctive transforms lies in the assumption that a sufficiently small block can be considered smooth, meanwhile, guaranteeing the compression of the luma and chroma signals at the cost of a small overhead for coding the description of the designed luma graph. In addition, a color image compression framework based on the CCGBT is designed for comparing DCT on the classic JPEG codec. The proposed method benefits from its flexible transform block design on arbitrary sizes to exploit image content better than the fixed transform. The experimental results show that the unified graph-based transform outperforms conventional DCT, while close to discrete wavelet transform on JPEG2000 at high bit-rates.

Learning Content-Weighted Pseudocylindrical Representation for 360° Image Compression.

Learned 360° image compression methods using equirectangular projection (ERP) often confront a non-uniform sampling issue, inherent to sphere-to-rectangle projection. While uniformly or nearly uniformly sampling representations, along with their corresponding convolution operations, have been proposed to mitigate this issue, these methods often concentrate solely on uniform sampling rates, thus neglecting the content of the image. In this paper, we urge that different contents within 360° images have varying significance and advocate for the adoption of a content-adaptive parametric representation in 360° image compression, which takes into account both the content and sampling rate. We first introduce the parametric pseudocylindrical representation and corresponding convolution operation, upon which we build a learned 360° image codec. Then, we model the hyperparameter of the representation as the output of a network, derived from the image's content and its spherical coordinates. We treat the optimization of hyperparameters for different 360° images as distinct compression tasks and propose a meta-learning algorithm to jointly optimize the codec and the metaknowledge, i.e., the hyperparameter estimation network. A significant challenge is the lack of a direct derivative from the compression loss to the hyperparameter network. To address this, we present a novel method to relax the rate-distortion loss as a function of the hyperparameters, enabling gradient-based optimization of the metaknowledge. Experimental results on omnidirectional images demonstrate that our method achieves state-of-the-art performance and superior visual quality.

Fast and High-Performance Learned Image Compression With Improved Checkerboard Context Model, Deformable Residual Module, and Knowledge Distillation.

Deep learning-based image compression has made great progresses recently. However, some leading schemes use serial context-adaptive entropy model to improve the rate-distortion (R-D) performance, which is very slow. In addition, the complexities of the encoding and decoding networks are quite high and not suitable for many practical applications. In this paper, we propose four techniques to balance the trade-off between the complexity and performance. We first introduce the deformable residual module to remove more redundancies in the input image, thereby enhancing compression performance. Second, we design an improved checkerboard context model with two separate distribution parameter estimation networks and different probability models, which enables parallel decoding without sacrificing the performance compared to the sequential context-adaptive model. Third, we develop a three-pass knowledge distillation scheme to retrain the decoder and entropy coding, and reduce the complexity of the core decoder network, which transfers both the final and intermediate results of the teacher network to the student network to improve its performance. Fourth, we introduce L1 regularization to make the numerical values of the latent representation more sparse, and we only encode non-zero channels in the encoding and decoding process to reduce the bit rate. This also reduces the encoding and decoding time. Experiments show that compared to the state-of-the-art learned image coding scheme, our method can be about 20 times faster in encoding and 70-90 times faster in decoding, and our R-D performance is also 2.3% higher. Our method achieves better rate-distortion performance than classical image codecs including H.266/VVC-intra (4:4:4) and some recent learned methods, as measured by both PSNR and MS-SSIM metrics on the Kodak and Tecnick-40 datasets.

Compressive Sensing Based Image Codec With Partial Pre-Calculation

Compressive Sensing Based Image Codec With Partial Pre-Calculation

Sketch Assisted Face Image Coding for Human and Machine Vision: A Joint Training Approach

Image coding is one of the most fundamental techniques and is widely used in image/video processing and multimedia communications. Current image coding methods are mainly human-oriented, and the visual quality is always unsatisfactory, especially at low bitrates. Moreover, the recent emergence of machine vision goes beyond the scope of current coding. With these considerations, we proposed a sketch assisted face image coding for human and machine vision by a joint training approach. In the proposed approach, we design a new feature representation: a color sketch, which aims to satisfy both low-frequency features of human vision and high-frequency features of machine analysis. Then, we present a novel end-to-end image codec framework with joint training that consists of three models: an image-to-image translation module, a coding module, and a two-stage reconstruction module. Specifically, the input image is first translated into the edge map with the Canny edge as the auxiliary label to merely preserve the structure information. Afterward, the backpropagation from reconstruction module guides the edge map to increase or decrease the information through joint training, which results in the generation of color sketch. Then, the generated sketch is compressed into the bitstream and decompressed back to a sketch in the coding module. Finally, the decompressed sketch is reconstructed to support the machine and human tasks, respectively. In this way, the color sketch is designed to bridge the gap between human and machine vision, and the joint training strategy helps to adjust the low-frequency information in the sketch. The experimental results on challenge datasets demonstrate that our proposed algorithm offers 40.9%-86.6% bitrate savings on machine vision and is comparable to state-of-the-art image coding methods on human vision.

A Universal Optimization Framework for Learning-based Image Codec

Recently, machine learning-based image compression has attracted increasing interests and is approaching the state-of-the-art compression ratio. But unlike traditional codec, it lacks a universal optimization method to seek efficient representation for different images. In this paper, we develop a plug-and-play optimization framework for seeking higher compression ratio, which can be flexibly applied to existing and potential future compression networks. To make the latent representation more efficient, we propose a novel latent optimization algorithm to adaptively remove the redundancy for each image. Additionally, inspired by the potential of side information for traditional codecs, we introduce side information into our framework, and integrate side information optimization with latent optimization to further enhance the compression ratio. In particular, with the joint side information and latent optimization, we can achieve fine rate control using only single model instead of training different models for different rate-distortion trade-offs, which significantly reduces the training and storage cost to support multiple bit rates. Experimental results demonstrate that our proposed framework can remarkably boost the machine learning-based compression ratio, achieving more than 10% additional bit rate saving on three different representative network structures. With the proposed optimization framework, we can achieve 7.6% bit rate saving against the latest traditional coding standard VVC on Kodak dataset, yielding the state-of-the-art compression ratio.

Benchmarking Lossless Still Image Codecs: Perspectives on Selected Compression Standards From 1992 through 2022

Benchmarking Lossless Still Image Codecs: Perspectives on Selected Compression Standards From 1992 through 2022

Memory-Efficient Fixed-Length Representation of Synchronous Event Frames for Very-Low-Power Chip Integration

The new event cameras are now widely used in many computer vision applications. Their high raw data bitrate levels require a more efficient fixed-length representation for low-bandwidth transmission from the event sensor to the processing chip. A novel low-complexity lossless compression framework is proposed for encoding the synchronous event frames (EFs) by introducing a novel memory-efficient fixed-length representation suitable for hardware implementation in the very-low-power (VLP) event-processing chip. A first contribution proposes an improved representation of the ternary frames using pixel-group frame partitioning and symbol remapping. Another contribution proposes a novel low-complexity memory-efficient fixed-length representation using multi-level lookup tables (LUTs). Complex experimental analysis is performed using a set of group-size configurations. For very-large group-size configurations, an improved representation is proposed using a mask-LUT structure. The experimental evaluation on a public dataset demonstrates that the proposed fixed-length coding framework provides at least two times the compression ratio relative to the raw EF representation and a close performance compared with variable-length video coding standards and variable-length state-of-the-art image codecs for lossless compression of ternary EFs generated at frequencies bellow one KHz. To our knowledge, the paper is the first to introduce a low-complexity memory-efficient fixed-length representation for lossless compression of synchronous EFs, suitable for integration into a VLP event-processing chip.

Improving Image Compression With Adjacent Attention and Refinement Block

Recently, learned image compression algorithms have shown incredible performance compared to classic hand-crafted image codecs. Despite its considerable achievements, the fundamental disadvantage is not optimized for retaining local redundancies, particularly non-repetitive patterns, which have a detrimental influence on the reconstruction quality. This paper introduces the autoencoder-style network-based efficient image compression method, which contains three novel blocks, i.e., adjacent attention block, Gaussian merge block, and decoded image refinement block, to improve the overall image compression performance. The adjacent attention block allocates the additional bits required to capture spatial correlations (both vertical and horizontal) and effectively remove worthless information. The Gaussian merge block assists the rate-distortion optimization performance, while the decoded image refinement block improves the defects in low-resolution reconstructed images. A comprehensive ablation study analyzes and evaluates the qualitative and quantitative capabilities of the proposed model. Experimental results on two publicly available datasets reveal that our method outperforms the state-of-the-art methods on the KODAK dataset (by around 4dB and 5dB) and CLIC dataset (by about 4dB and 3dB) in terms of PSNR and MS-SSIM.

Accelerating BPC-PaCo through Visually Lossless Techniques

Fast image codecs are a current need in applications that deal with large amounts of images. Graphics Processing Units (GPUs) are suitable processors to speed up most kinds of algorithms, especially when they allow fine-grain parallelism. Bitplane Coding with Parallel Coefficient processing (BPC-PaCo) is a recently proposed algorithm for the core stage of wavelet-based image codecs tailored for the highly parallel architectures of GPUs. This algorithm provides complexity scalability to allow faster execution at the expense of coding efficiency. Its main drawback is that the speedup and loss in image quality is controlled only roughly, resulting in visible distortion at low and medium rates. This paper addresses this issue by integrating techniques of visually lossless coding into BPC-PaCo. The resulting method minimizes the visual distortion introduced in the compressed file, obtaining higher-quality images to a human observer. Experimental results also indicate 12% speedups with respect to BPC-PaCo.

Multimedia stream hashing: A forensic method for content verification.

Forensic examination of digital audio, video, and images frequently requires transforming multimedia data from one format to another. The transformative activity may cause changes to the administrative elements of the file but leave the multimedia streams unchanged and intact. However, the forensic science community has a method knowledge gap in accurately determining if the multimedia streams changed or remained the same in the transformative processes. This paper illustrates the practical use of multimedia stream hashing as a forensic method for verifying multimedia content. A universal stream hashing tool decodes the multimedia stream data at rest in a file container. Subsequently, it calculates the data stream hash using reference video, audio, and image codecs. This paper illustrates that the multimedia stream hashing method can accurately confirm the integrity of digital images, videos, and audio following transmission, transcoding, or re-containerization. Our findings confirmed that stream hashing could accurately detect changes in multimedia streams during transcoding. Furthermore, the stream hashing method can also accurately detect matching multimedia streams. In addition, this paper verified the forensic use of the multimedia stream hash method while establishing the error rate for its use. The hash algorithms used in stream hashing have zero false negative rates by design. However, the false positive (error rate) is extremely low and depends on hashing algorithm. Finally, we recommend the forensic science community adopt the multimedia stream hashing method as an initial testing method. The method can verify a multimedia stream's conversion (transcoding) from one codec to another using FFmpeg.

Joint image compression and denoising via latent-space scalability

When it comes to image compression in digital cameras, denoising is traditionally performed prior to compression. However, there are applications where image noise may be necessary to demonstrate the trustworthiness of the image, such as court evidence and image forensics. This means that noise itself needs to be coded, in addition to the clean image itself. In this paper, we present a learning-based image compression framework where image denoising and compression are performed jointly. The latent space of the image codec is organized in a scalable manner such that the clean image can be decoded from a subset of the latent space (the base layer), while the noisy image is decoded from the full latent space at a higher rate. Using a subset of the latent space for the denoised image allows denoising to be carried out at a lower rate. Besides providing a scalable representation of the noisy input image, performing denoising jointly with compression makes intuitive sense because noise is hard to compress; hence, compressibility is one of the criteria that may help distinguish noise from the signal. The proposed codec is compared against established compression and denoising benchmarks, and the experiments reveal considerable bitrate savings compared to a cascade combination of a state-of-the-art codec and a state-of-the-art denoiser.

A no-reference perceptual image quality assessment database for learned image codecs

The drastic growth of research in image compression, especially deep learning-based image compression techniques, poses new challenges to objective image quality assessment (IQA). Typical artifacts encountered in the emerging image codecs are significantly different from that produced by traditional block-based codecs, leading to inapplicability of the existing objective IQA algorithms. Towards advancing the development of objective IQA algorithms for recent compression artifacts, we built a learning-based compressed image quality assessment (LCIQA) database involving traditional block-based image codecs, hybrid neural network based image codecs, convolutional neural network based and generative adversarial network (GAN) based end-to-end optimized image coding approaches. Our study confirms the statistical difference and human perception difference between reconstructions of learned compression and traditional block-based compression. We propose a two-step deep learning model for learning-based compressed image quality assessment. Extensive experiments on LCIQA database demonstrate that our proposed model performs better than other counterparts on learning-based compressed images, especially on GAN compressed images, and achieves competitive performance to the state-of-the-art IQA metrics on traditional compressed images.

A novel framework for image-to-image translation and image compression

Data-driven paradigms using machine learning are becoming ubiquitous in image processing and communications. In particular, image-to-image (I2I) translation is a generic and widely used approach to image processing problems, such as image synthesis, style transfer, and image restoration. At the same time, neural image compression has emerged as a data-driven alternative to traditional coding approaches in visual communications. In this paper, we study the combination of these two paradigms into a joint I2I compression and translation framework, focusing on multi-domain image synthesis. We first propose distributed I2I translation by integrating quantization and entropy coding into an I2I translation framework (i.e. I2Icodec). In practice, the image compression functionality (i.e. autoencoding) is also desirable, requiring to deploy alongside I2Icodec a regular image codec. Thus, we further propose a unified framework that allows both translation and autoencoding capabilities in a single codec. Adaptive residual blocks conditioned on the translation/compression mode provide flexible adaptation to the desired functionality. The experiments show promising results in both I2I translation and image compression using a single model.

An Image Importance Partition-Based Compression Method for COVID-19 Computed Tomography Scan

As the coronavirus disease 2019 (COVID-19) spreads around the world, industrial automated medical diagnosis systems have been developed, which complete a large amount of medical diagnosis work through computed tomography (CT) images. In these systems, how to quickly store and transmit such a large amount of CT image information has important research significance. In this article, a more targeted COVID-19 chest CT image codec is proposed to make image data not only occupy less space but also have higher image quality. First, the bilateral lung contours are extracted to calculate the position information of the region of interest (ROI). Then, a CT image is classified into four types of nonuniform image blocks according to the characteristics of COVID-19 chest CT images and ROI position information. Next, a series of new transformations are proposed for more efficient transform coding. Finally, a flexible quantization strategy is proposed for the adaptive quantization part. In the experiments, the proposed method is superior to some of the existing methods with similar computational complexity. At the same bit rate, it significantly improves the image quality. This means that chest CT images can still be used for disease diagnosis while taking up less space. In addition, because of the low computational complexity of the proposed method, it can be more easily embedded into the CT equipment with low computational power.